PROCEEDINGS, Eleventh Workshop OD Geothermal Reservoir Engineering
Stanford University, Stanford, California, January 21-23, 1986
SGP-TR-93
HOT SPRING MONITORING AT
LASSEN VOLCANIC NATIONAL PARK, CALIFORNIA 1983-1985
Michael L. Sorey
U . S . Geological Survey
Menlo Park, California
ABSTRACT
Data collected on several occasions between 1983 and 1985 as part of
a hydrologic monitoring program by the
U.S. Geological Survey permit.preliminary estimation of the natural
varability in the discharge characteristics of hydrothermal features in
Lassen Volcanic National Park and the
Lassen XGRA in northern California.
The total rate of discharge of highchloride hot springs along Mill Creek
and Canyon Creek in the Lassen XGRA
has averaged 20.9 2 1.7 L / s , based on
seven measurements of the flux of
chloride in these streams. Measured
chloride flux does not appear to
increase with streamflow during the
spring-summer snowmelt period, as
observed at Ye1 lowstone and Long
Val ley Caldera. The corresponding
fluxes of arsenic in Mill Creek and
Canyon Creek decrease within distances
of about 2 km downstream from the hot
springs by approximately 30%, most
likely due to chemical absorption on
streambed sediments. Within Lassen
Volcanic National Park, measurements
of sulfate flux in streams draining
steam-heated thermal features at
Sulphur Works and Bumpass Hell have
averaged 7.5 +, 1.0 and 4.0 5 1.5 g/sl
respectively. Calculated rates of
steam upf low containing,dissolved H S
to supply these sulfate fluxes arek.8
kg/s at Sulphur Works and 1.0 kg/s at
Bumpass Hell.
INTRODUCTION
This paper summarizes data
collected in a hydrologic monitoring
program conducted by the U.S.
Geological Survey between 1983 and
1985 in the vicinity of Lassen
Volcanic National Park (LVNP). The
area of study is located in northcentral California at the southern end
of the Cascade Range and includes LVNP
and the adjacent Lassen Known
Geothermal Resource Area (XGRA, fig.
1). The monitoring program has been
funded by the National Park Service in
an effort to determine the natural
level of variability in the discharge
characteristics of surficial hydrothermal features in this area. Such
information may be needed to assess
the impacts of future geothermal
development on the Lassen hydrothermal
system. An Environmental Impact
Statement is currently being prepared
for geothermal leasing of federal
lands adjacent to LVNP.
Late Pliocene to Holocene
volcanic rocks in the Lassen region
were extruded from several long-lived
volcanic centers, each producing composite cones of andesite to rhyolite
lavas and pyroclastics (Clynne, 1984,
1985). A silica magma chamber associated with the Lassen volcanic center
(0.6 m.y. to present) provides heat
for the present-day hydrothermal
system which occurs primarily within
the southernhalf of L V N P a n d t h e
adjacent KGRA (fig. 2).
Surface thermal discharge
includes fumaroles and steam-heated
hot springs at relatively high elevations within the Park and highchloride hot springs at relatively low
elevation in the KGRA. Geochemical
evidence and temperature measurements
in a 1.2 km deep well near Terminal
Geyser indicate that the Lassen hydrothermal system is liquid-dominated but
includes a parasitic vapor-dominated
region associated with the central
upf low zone beneath Bumpass He1 1
(Muffler et al., 1982, Sorey and
Ingebritsen, 1984, Ingebritsen and
SOrey;, 1985). Zones of lateral outflow of hot water derived from this
upflow zone trend southward and
southeastward, extending to discharge
points within the XGRA. Thermal fluid
discharge occurs in high chloride hot
springs at Growler and Morgan Hot
Springs and in diluted low chloride
springs at Domingo Springs (fig. 2).
Additional thermal fluid may flow in
shallow aquifers south of Morgan Hot
Springs and may discharge as seeps
~
-141-
into Willow Creek and Warner Creek
(Sorey and Ingebritsen, 1984).
in Domingo Springs. In results shown
in the next section, thermal fluid
discharge at Domingo Springs is calculated from the measured flow of the
springs by assuming that the chloride
concentration of the thermal component
is equal t o t h a t measured inwater
produced from the well at Terminal
Geyser (2180 mg/L). This, in turn, is
based on the assumption that some or
all of the thermal water flowing
southeast in the lateral outflow zone
beneath Terminal Geyser eventual ly
discharges in Domingo Springs.
THE MONITORING PROGRAM
The monitoring program described
in this paper includes periodic
streamflow measurements and chemical
sampling at several sites in Mill
Creek, Canyon Creek, and at Domingo
Springs, and similar measurements in
streams draining steam-heated thermal
features at Sulphur Works and Bumpass
Hell (fig. 2). Such data permit calculation of hot spring discharge
within the KGRA from measurements of
the flux of chemical elements contributed to the streams by hot spring
waters. With few exceptions, the flow
of hot springs in the Lassen area
cannot be gaged directly and much of
the thermal water occurs as streambed
seepage. Hence thermal fluid
discharge must be calculated from
chemical flux measurements.
Chemical flux measurements in the
Lassen KGRA have been made for three
conservative thermal fluid elements
C1, B, and As. The precision of this
technique is enhanced by the
relatively high concentration of these
elements in the hot springs (table 1).
Such concentrations may be indicative
of a fluid component of marine origin
within the Lassen hydrothermal system
(Sorey and Ingebritsen, 1984;
Thompson, 1985). Although the fraction of hot-spring water in Mill Creek
and Canyon Creek at the measuring
sites ranges from about 0.5% to 5%
during the year, concentrations of C1,
B, and As remain well above detection
limits. As noted below, however, a
significant portion of the arsenic
contributed to these streams by the
hot springs appears to be taken out of
solution by adsorption on streambed
sediments, making chloride flux
measurements the most reliable for
calculating hot spring discharge.
-
At Domingo Springs, water that
flows from several vents at temperatures near 9OC is channeled through a
culvert. Streamflow measurements and
chemical samples have been taken below
the discharge end of the culvert.
Although the measured ‘chloride concentrations in this water (11-19 mg/L)
are signifcantly lower than chloride
concentrations in Growler and Morgan
Hot springs, they are approximately 10
times higher than chloride concentrations inother coldsprings i n t h e
KGRA. This indicates that some
component of thermal water from the
Lassen hvdrothermal svstem discharqes
-142-
The rate of discharge of fluids
from thermal areas within LVNP cannot
be determined by the same type of
chemical flux measurements because of
the loss of steam in fumaroles a n d t h e
lack of conservative elements such as
C1 in the steam-heated hot springs.
The discharge characteristics of
individual vents in such areas may
undergo considerable fluctuation due
to seasonal and annual hydrologic
variations. Pub1 ished and unpublished
data on the temperature and gas
concentration of selected steam vents
at thermal areas in LVNP document some
of these changes that have occurred
since 1975 (Muffler et al., 1982;
Truesdell et. al., 1983; Janik,
written communication, 1986). For the
purposes of delineating changes in the
rate of upflow of steam and gas from
underlying vapor-dominated zones,
measurements of the total heat output
from each thermal area would be most
useful. Such measurements, although
difficult to obtain, have been
successfully made at Wairakei, New
Zealand, to quantify changes in steam
upflow induced by geothermal development (Allis, 1981). Total heat output
determinations require measurements of
heat loss by advection in hot springs
and fumaroles, by conduction, and by
evaporation.
Data collected during the present
monitoring program on the flux of
sulfate in streams draining thermal
areas at Sulphur Works and Bumpass
Hell provide a preliminary basis on
which to assess the natural variability in heat discharge at these
areas. Some or a l l of this sulfate
flux must come from near-surface oxidation of H2S which originates along
with steam upflow from underlying
vapor-dominated zones. If the ratio
of H2S to steam in such upflow is
known, the sulfate flux measurements
provide an estimate of the fraction of
the total steam upflow that is
condensed in the hot spring waters in
these areas. Such measurements were
originally suggested by A.H. Truesdell
(written communication, 1984), based
on the work of Giggenbach and Stewart
(1982). Complicating factors include
the possibility that sulfate is also
contributed to the streams draining
these areas from products of rock
weathering and that some H2S reacts
with near-surface rocks to form pyrite
and sulfur.
Measurements of sulfate
flux and the associated steam upflow
reported here are compared with
estimates of steam upflow made by
other means and by other investigators
to evaluate the effects of such
complicating factors.
accompanying the spring-summer runoff
period.
Results of hot spring discharge
calculations based on chloride flux
determinations for each thermal area
in the KGRA are listed in table 3.
The total hot spring flow along Mill
Creek (from Morgan Hot Springs) and
Canyon Creek (from Growler Hot
Springs) has average 20.9 L / s with a
standard deviation of 1.7 L/s. Note
that the total for Morgan Hot Springs
is calculated from the thermal
component in Mill Creek at Highway 36
(17.2 L / s from table 2) less the
thermal component in Canyon Creek
below Growler Hot Springs (2.19 L / s
from table 3) plus the thermal component (3.7 L/s) in a tributary to Mill
Creek at Highway 3 6 t h a t i s also
derived from hot springs at Morgan Hot
Springs. At Doming0 Springs, chloride
flux measurements indicate an average
thermal component of 0.93 2 0.18 L/s
with chloride concentration equal to
that measured in the well at Terminal
Geyser.
THERMAL WATER DISCHARGE IN LASSEN KGRA
The rate of discharge of thermal
water from hot springs along Mill
Creek and Canyon Creek can be estimated from the difference in the flux
of chloride at sites above and below
the hot springs divided by the
chloride concentration in the spring
waters. Data f o r t h e Site o n M i l l
Creek at Highway 36 (table 2)
exemplify the way this is done. Calculated values of hot spring flow
(thermal water component) are based on
an average chloride concentration of
2 3 0 0 m g / L i n h o t s p r i n g s abovethis
site. The average spring flow is
calculated as 17.2 L/s with a standard
deviation of 1.2 L/s. Thermal water
discharge at Morgan Hot Springs and
Growler Hot Springs contributes to
this total.
Measured fluxes of arsenic in
Mill Creek and Canyon Creek are significantly lower than values calculated
from estimates of the thermal-fluid
component in these streams (based on
measured chloride fluxes). This is
not true for the measured boron
fluxes, which agree with the calculated fluxes. Data for Mill Creek at
Highway 36 (table 4) indicate that 1039% of the hot-spring derived arsenic
is lost between the springs and this
measurement site over a distance of
2.5 km. The most likely explanation
for this loss is chemical adsorption
on streambed sediments. Similar c a l culations for Canyon Creek show
arsenic losses of 19-50% over a
distance of 1.6 km.
Differences between calculated
values of hot spring discharge at this
site and other sites within the KGRA
between the seven sampling visits
during the 1983-1985 period could be
caused by several factors. These
include errors in streamflow measurements and chloride determinations,
loss (or gain) of stream water to the
shallow ground-water system below the
hot spring areas, and actual
differences in hot spring discharge.
Data from chemical analysis on
duplicate water samples and subjective
ratings of stream gaging accuracy
suggest that the accuracy of the flux
measurements at this site is t 10%.
Thus, most if not a l l ofthevariation
in calculated spring flow at this site
could be due to errors in streamflow
and chloride concentration determinations. There does not appear to be a
consistent increase in chloride flux
at this site (or any of the other
sites in the KGRA) with increasing
streamflow.
Such a correlation has
been shown for hot spring areas at
Yellowstone National Park (Norton and
Friedman, 1985) and Long Valley
Caldera (Farrar, et al., 1985). In
those areas the correlation has been
attributed to a release of chloride
from storage within drainage basins
SULFATE DISCHARGE IN LVNP
Measurements of sulfate flux in
streams draining steam-heated thermal
features at Sulphur Works (SW) and
Bumpass Hell (BH) in LVNP are reported
in table 5. Measurements at SW have
been made at four culverts through
which branches of West Sulphur Creek
flow. Data in table 5 apply to two of
these branches which drain active
thermal features: sulfate discharge in
the other two branches is contributed
entirely by erosion of sulfate-rich
surficial deposits. The total hotspring derived sulfate flux at SW
averages 7.5 5 1.0 q/s, and the corresponding total for BH averages 4.0 +
-143-
1.5 L/s. At both areas, the reported
sulfate flux does not vary significantly with streamflow, indicating
that the sulfate contribution from
nonthermal sources is minimal.
REFERENCES
Clynne, M.A., 1984, Stratigraphy and
major-element geochemistry of the
Lassen Volcanic Center, California:
U.S. Geological Survey Open-File
Report 84-224, 168 p.
As discussed previously, the sulfate flux estimates can be converted
to estimates of steam condensed in the
hot springs, assuming the sulfate is
derived from oxidation of H2S and that
the ratio of oxidized H2S to condensed
steam is the same as the H2S to steam
ratio in the vapor-dominated reservoir.
Data from Muffler et. al.
(1982) based on analyses of fumarolic
gases at BH indicate that the steam to
H2S molar ratio in the vapor-dominated
reservoir is 1290. The corresponding
weight ratio of s t e a m t o S 0 4 w o u l d b e
242. Thus the average sulfate fluxes
at BH and SW noted above convert to
1.1 kg/s and 1.8 kg/s of condensed
steam, respectively. For comparison,
the rate of steam condensation required to provide the advective heat
loss in the streams draining these
areas is approximately twice the
values calculated from the sulfate
measurements. This suggests that the
actual ratio of condensed steam to
oxidized H2S is at least twice the
steam t o H 2 S ratio in the vapordominated reservoir. The higher steam
to H2S ratios implied by the advective
heat loss rates may be accounted for
by near surface loss of H2S to form
pyrite and sulfur deposits and by
surficial discharge of H2S in gas
vents and seeps.
Clynne, M.A., 1985, Quaternary volcanism
in the Southern Cascade Range,
California: in Proceedings of the
Workshop on GGthermal Resources of
the Cascade Range, ed. Guffant, M.,
and Muffler ,.S.
L.
U:.P.J
Geological Survey Open-File Report
85-521, p. 24-30.
Farrar, C.D. , Sorey, M.L. , Rostaczer,
S.A., Janik, C.J., Mariner, R.H.,
Winnett, T.L., and Clark, M.A.,
1985, Hydrologic and geochemical
monitoring in Long Valley caldera,
Mono County, California 1982-1984:
U.S. Geological Survey Water
Resources Investigations Report 854183, 137 p.
Friedman, J.D. and Frank, D., 1978,
Thermal surveillance of active
volcanoes using Landsat-1 Data
collection System: Part 4, Lassen
volcanic region: National Technical
Information Services, NTIS N78
23499/LL, 46 p.
Giggenbach, W.F. and Stewart, M.X., 1982,
Processes controlling the isotopic
composition of steam and water
discharge from steam vents and
steam-heated pools in geothermal
areas: Geothermics, v. 11, p. 71-
Estimates of the total rate of
steam upflow at Bumpass Hell can also
be obtained from the calculations of
total heat flow at this area discussed
by Friedman and Frank (1978). Their
data, based on ground temperature
measurements and meterological calculations, together with the advective
heat loss estimates from the current
monitoring program, indicate a total
heat 10s of 36 M w o v e r a n a r e a of
0.045 km 3 at Bumpass Hell (fig. 3).
This total does not include heat loss
in fumarolic steam discharge. The
rate of steam upflow required to maintain this heat flow is approximately
13 kg/s, or ten times flow rate
suggested by the sulfate flux measurements. Additional study and field
observations will be required to
verify this heat flow estimate and to
delineate more clearly the processes
responsible for these differences in
steam upf l o w estimates.
80.
Ingebritsen, S.E. and Sorey, M.L., 1985,
A quantitative analysis of the
Lassen hydrothermal system, northcentral California, Water Resources
Research, v. 21, no. 6, p. 853-868.
Muffler, L.J.P., Nehring, N.L.,
Truesdell , A.H. , Janik, C. J. ,
Clynne, M.A., and Thompson, J.M.,
1982, The Lassen geothermal system:
Proceedings of the Pacific
Geothermal Conference, Auckland, New
Zealand, p. 349-356.
Norton, D.R., and Friedman, I., 1985,
Chloride flux out of Yellowstone
National Park: Journal of
Volcanology and Geothermal Research,
V. 26, p. 231-250.
Sorey, M.L., and Ingebritsen, S.E., 1984,
Quantitative analysis of the
hydrothermal system in Lassen
Volcanic National Park and Lassen
Known Geothermal Resource Area:
U.S. Geological Survey Water
-144-
Resources Investigations Report 844278, 80 p.
Thompson, J.M., 1985, Chemistry of
thermal and nonthennal springs in
the vicinity of Lassen Volcanic
National Park: Journal of
Volcanology and Geothermal Energy,
V. 25, p. 81-104.
Truesdell, A.H., Mazor, E., and Nehring,
N.L., 1983, The origin of thermal
fluids at Lassen Volcanic National
Park: Evidence from noble and
reactive gas abundances: Geothermal
Resources Council Transactions, v.
7, p. 343-348.
Table 1.
Average concentrations of selected ions in thermal
waters from the Lassen area, based on analytical
results from the U.S. Geological Survey Central
Laboratory, Arvada, Colorado on samples collected
1983
1985 (except for data for the Terminal Geyser
well which are from Thompson, 1985).
-
Location
Growler Hot Springs
94
2500
87
11
Morgan Hot Springs
94
2300
80
10
175
2180
62
10
9
15
Terminal Geyser Well
Doming0 Springs
Table 2.
0.5
0.03
Data on streamflow and chloride concentration in
Mill Creek at Highway 36 and corresponding calculated
values of chloride flux and thermal-water component.
-145-
T a b l e 3. Average values of thermal-water
discharge a t thermal areas i n t h e L a s s e n
XGRA, based on chloride f l u x measurements
between 1 9 8 3 and 1985. S t a n d a r d deviat i o n s f o r each s e t of v a l u e s shown w i t h 2.
S p r i n g area
T a b l e 4.
T h e r m a l water
discharge ( L / s )
C a l c u l a t e d l o s s e s of h o t - s p r i n g d e r i v e d b o r o n a n d a r s e n i c i n
M i l l Creek a b o v e Highway 3 6 , b a s e d on h o t s p r i n g d i s c h a r g e
computed f r o m m e a s u r e d c h l o r i d e f l u x e s ( s e e t a b l e 2 ) .
6/84
4871
0.18
0.12
32
1.44
1.46
-1
8/84
1170
0.16
0.10
39
1.27
1.17
8
11/84
1034
0.18
0.12
31
1.44
1.34
7
7/85
793
0.16
0.14
10
1.27
1.31
-3
10185
1025
0.20
0.16
18
1.61
1.54
4
T a b l e 5.
Date
S u l f a t e f l u x m e a s u r e m e n t s a t S u l p h u r Works a n d Bumpass H e l l .
S u l p h u r Works
SO4
Streamflow
(L/S)
(mg/L)
so4
flux
(g/s)
Bumpass Hell
SO4
Streamflow
(L/S)
(mg/L)
so4
flux
(g/s)
..........................................................................
8/83*
280
13
3.6
8/84
54
118
6.4
11/84
52
148
7.7
5/85*
72
60
4.3
7/85
52
117
6.1
10185
39
203
7.9
57
68
3.9
5.4
440
2.4
6.2
850
5.3
180
6.5
4.0
740
3.0
2.8
990
2.8
36
..........................................................................
-
*Measurements made o n l y a t " c e n t r a l " c u l v e r t
SO4 flux a p p l i e s
o n l y t o " c e n t r a l " c u l v e r t ; measurements on a l l o t h e r d a t e s made a t
" c e n t r a l ' c u l v e r t and "east" c u l v e r t
SO4 f l u x a p p l i e s t o t h e t o t a l
f o r both culverts.
-
-146-
LEASE AREA
1000 k m 2
Evaporation l Q MW
Conduction 1 1
Total
Advection
Heat ~6~~
6
A
I
KGRA
1 kg18
I
BUMPASS HELL
Figure 1. Map showing location of the Lassen
study area i n north-central California, which
includes Lassen (Volcanic National) Park, Lassen
KGRA, and Thousand Lakes Wilderness Area (crosshatched pattern i n upper l e f t ) . All o r p a r t of
the study area excluding the cross-hatched areas
may be leased f o r geothermal development by the
Bureau of Land Management during 1986.
/4J
Figure 3. Diagram of Bumpass Hell thermal area
depicting the average value of s u l f a t e flux i n
Bumpass Creek ( 4 g / s ) and the corresponding r a t e
of steam condensation ( 1 kg/s) assuming a steam
t o H2S molar r a t i o of 1290. Also shown a r e heat
flow estimates based on the work of Friedman and
Frank (1978) and the r e s u l t s of t h i s study and
the corresponding r a t e of steam upflow required
t o supply a t o t a l heat flow of 36 MW (13 kg/s).
40'35'
L A S E N VOLCANIC
NATIONAL PARU
I
I
10'25
wm
1zlom'
0
13 k g l e
IO km
Figure 2 . Map showing areas o f thermal-fluid discharge
and surface drainage i n the Lassen region. Areas w i t h
fumaroles and steam-heated hot springs a r e shown as t r i angles ( B H = Bumpass Hell, LHSV = L i t t l e Hot Springs
Valley, S
W = Sulphur Works, D K = Devils Kitchen, DB =
Drakesbad, BSL = Boiling Springs Lake, TG = Terminal
Geyser). Areas w i t h high-chloride thermal water discharge
a r e shown as dots (GHS = Growler Hot S p r i n g s , MHS =
Morgan Hot Springs, DS = Doming0 Springs). Short l i n e s
crossing streams a t r i g h t angles s i g n i f y s i t e s where
chemical flux measurements have been made f o r estimation
of thermal-fluid discharge.
-147-